Generating an image from a given text description has two goals: visual realism and semantic consistency. Although significant progress has been made in generating high-quality and visually realistic images using generative adversarial networks, guaranteeing semantic consistency between the text description and visual content remains very challenging. In this paper, we address this problem by proposing a novel global-local attentive and semantic-preserving text-to-image-to-text framework called MirrorGAN. MirrorGAN exploits the idea of learning textto-image generation by redescription and consists of three modules: a semantic text embedding module (STEM), a global-local collaborative attentive module for cascaded image generation (GLAM), and a semantic text regeneration and alignment module (STREAM). STEM generates word-and sentence-level embeddings. GLAM has a cascaded architecture for generating target images from coarse to fine scales, leveraging both local word attention and global sentence attention to progressively enhance the diversity and semantic consistency of the generated images. STREAM seeks to regenerate the text description from the generated image, which semantically aligns with the given text description. Thorough experiments on two public benchmark datasets demonstrate the superiority of Mirror-GAN over other representative state-of-the-art methods.
A recent ''third wave'' of neural network (NN) approaches now delivers state-ofthe-art performance in many machine learning tasks, spanning speech recognition, computer vision, and natural language processing. Because these modern NNs often comprise multiple interconnected layers, work in this area is often referred to as deep learning. Recent years have witnessed an explosive growth of research into NN-based approaches to information retrieval (IR). A significant body of work has now been created. In this paper,
In this paper, we propose a highly accurate inpainting algorithm which reconstructs an image from a fraction of its pixels. Our algorithm is inspired by the recent progress of non-local image processing techniques following the idea of 'grouping and collaborative filtering'. In our framework, we first match and group similar patches in the input image, and then convert the problem of estimating missing values for the stack of matched patches to the problem of low-rank matrix completion, and finally obtain the result by synthesizing all the restored patches. In our algorithm, how to accurately perform patch matching process and solve the low-rank matrix completion problem are key points. For the first problem, we propose a robust patch matching approach, and for the second task, the alternating direction method of multipliers is employed. Experiments show that our algorithm has superior advantages over existing inpainting techniques. Besides, our algorithm can be easily extended to handle practical applications including rendering acceleration, photo restoration and object removal.
Recently, research on the electrocatalytic CO 2 reduction reaction (eCO 2 RR) has attracted considerable attention due to its potential to resolve environmental problems caused by CO 2 while utilizing clean energy and producing high-value-added products. Considerable theoretical research in the lab has demonstrated its feasibility and prospect. However, industrialization is mandatory to realize the economic and social value of eCO 2 RR. For industrial application of eCO 2 RR, more criteria have been proposed for eCO 2 RR research, including high current density (above 200 mA cm −2 ), high product selectivity (above 90%), and long-term stability. To fulfill these criteria, the eCO 2 RR system needs to be systematically designed and optimized. In this review, recent research on eCO 2 RR for industrial applications is summarized. The review starts with focus on potential industrial catalysts in eCO 2 RR. Next, potential industrial products are proposed in eCO 2 RR. These products, including carbon monoxide, formic acid, ethylene, and ethanol, all have high market demand, and have shown high current density and product selectivity in theoretical research. Notably, the innovative components and strategy for industrializing the eCO 2 RR system are also highlighted here, including flow cells, seawater electrolytes, solid electrolytes, and a two-step method. Finally, some instructions and possible future avenues are presented for the prospects of future industrial application of eCO 2 RR.
One of the main challenges in ultrafast material science is to trigger phase transitions with short pulses of light. Here we show how strain waves, launched by electronic and structural precursor phenomena, determine a coherent macroscopic transformation pathway for the semiconducting-to-metal transition in bistable Ti3O5 nanocrystals. Employing femtosecond powder X-ray diffraction, we measure the lattice deformation in the phase transition as a function of time. We monitor the early intra-cell distortion around the light absorbing metal dimer and the long range deformations governed by acoustic waves propagating from the laser-exposed Ti3O5 surface. We developed a simplified elastic model demonstrating that picosecond switching in nanocrystals happens concomitantly with the propagating acoustic wavefront, several decades faster than thermal processes governed by heat diffusion.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.